From: Jiri Olsa <jolsa@redhat.com>
To: Andi Kleen <andi@firstfloor.org>
Cc: acme@kernel.org, jolsa@kernel.org, linux-kernel@vger.kernel.org
Subject: Re: Optimize perf stat for large number of events/cpus
Date: Wed, 20 Nov 2019 16:16:25 +0100 [thread overview]
Message-ID: <20191120151625.GG4007@krava> (raw)
In-Reply-To: <20191116055229.62002-1-andi@firstfloor.org>
On Fri, Nov 15, 2019 at 09:52:17PM -0800, Andi Kleen wrote:
> [v7: Address review feedback. Fix python script problem
> reported by 0day. Drop merged patches.]
>
> This patch kit optimizes perf stat for a large number of events
> on systems with many CPUs and PMUs.
>
> Some profiling shows that the most overhead is doing IPIs to
> all the target CPUs. We can optimize this by using sched_setaffinity
> to set the affinity to a target CPU once and then doing
> the perf operation for all events on that CPU. This requires
> some restructuring, but cuts the set up time quite a bit.
>
> In theory we could go further by parallelizing these setups
> too, but that would be much more complicated and for now just batching it
> per CPU seems to be sufficient. At some point with many more cores
> parallelization or a better bulk perf setup API might be needed though.
>
> In addition perf does a lot of redundant /sys accesses with
> many PMUs, which can be also expensve. This is also optimized.
>
> On a large test case (>700 events with many weak groups) on a 94 CPU
> system I go from
>
> real 0m8.607s
> user 0m0.550s
> sys 0m8.041s
>
> to
>
> real 0m3.269s
> user 0m0.760s
> sys 0m1.694s
>
> so shaving ~6 seconds of system time, at slightly more cost
> in perf stat itself. On a 4 socket system with the savings
> are more dramatic:
>
> real 0m15.641s
> user 0m0.873s
> sys 0m14.729s
>
> to
>
> real 0m4.493s
> user 0m1.578s
> sys 0m2.444s
>
> so 11s difference in the user visible set up time.
>
> Also available in
>
> git://git.kernel.org/pub/scm/linux/kernel/git/ak/linux-misc perf/stat-scale-10
>
> v1: Initial post.
> v2: Rebase. Fix some minor issues.
> v3: Rebase. Address review feedback. Fix one minor issue
> v4: Modified based on review feedback. Now it maintains
> all_cpus per evlist. There is still a need for cpu_index iteration
> to get the correct index for indexing the file descriptors.
> Fix bug with unsorted cpu maps, now they are always sorted.
> Some cleanups and refactoring.
> v5: Split patches. Redo loop iteration again. Fix cpu map
> merging for uncore. Remove duplicates from cpumaps. Add unit
> tests.
> v6: Address review feedback. Fix some bugs. Add more comments.
> Merge one invalid patch split.
> v7: Address review feedback. Fix python scripting (thanks 0day)
> Minor updates.
I posted another 2 comments, but other than that I think it's ok
I don't like it, but can't see a better way ;-) and the speedup
is really impressive
thanks,
jirka
next prev parent reply other threads:[~2019-11-20 15:16 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-16 5:52 Optimize perf stat for large number of events/cpus Andi Kleen
2019-11-16 5:52 ` [PATCH v7 01/12] perf pmu: Use file system cache to optimize sysfs access Andi Kleen
2019-11-16 5:52 ` [PATCH v7 02/12] perf affinity: Add infrastructure to save/restore affinity Andi Kleen
2019-11-16 5:52 ` [PATCH v7 03/12] perf cpumap: Maintain cpumaps ordered and without dups Andi Kleen
2019-11-16 5:52 ` [PATCH v7 04/12] perf evlist: Maintain evlist->all_cpus Andi Kleen
2019-11-16 5:52 ` [PATCH v7 05/12] perf evsel: Add iterator to iterate over events ordered by CPU Andi Kleen
2019-11-16 5:52 ` [PATCH v7 06/12] perf evsel: Add functions to close evsel on a CPU Andi Kleen
2019-11-16 5:52 ` [PATCH v7 07/12] perf stat: Use affinity for closing file descriptors Andi Kleen
2019-11-16 5:52 ` [PATCH v7 08/12] perf stat: Factor out open error handling Andi Kleen
2019-11-16 5:52 ` [PATCH v7 09/12] perf stat: Use affinity for opening events Andi Kleen
2019-11-20 15:07 ` Jiri Olsa
2019-11-20 22:31 ` Andi Kleen
2019-11-21 15:13 ` Jiri Olsa
2019-11-16 5:52 ` [PATCH v7 10/12] perf stat: Use affinity for reading Andi Kleen
2019-11-20 15:12 ` Jiri Olsa
2019-11-16 5:52 ` [PATCH v7 11/12] perf evsel: Add functions to enable/disable for a specific CPU Andi Kleen
2019-11-16 5:52 ` [PATCH v7 12/12] perf stat: Use affinity for enabling/disabling events Andi Kleen
2019-11-20 15:16 ` Jiri Olsa [this message]
-- strict thread matches above, loose matches on Subject: below --
2019-11-21 0:15 Optimize perf stat for large number of events/cpus Andi Kleen
2019-11-21 12:47 ` Andi Kleen
2019-11-21 14:32 ` Arnaldo Carvalho de Melo
2019-11-27 15:16 ` Arnaldo Carvalho de Melo
2019-11-27 15:43 ` Arnaldo Carvalho de Melo
2019-11-27 23:26 ` Andi Kleen
2019-11-28 0:01 ` Arnaldo Carvalho de Melo
2019-11-12 0:59 Andi Kleen
2019-11-07 18:16 Andi Kleen
2019-11-05 0:25 Andi Kleen
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20191120151625.GG4007@krava \
--to=jolsa@redhat.com \
--cc=acme@kernel.org \
--cc=andi@firstfloor.org \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.